The external validity of controlled clinical trials of psychotherapy for depression and anxiety: a naturalistic study

Abstract

Psychotherapy researchers have increasingly called for clinical practice and training to focus on empirically supported therapies tested in randomized controlled clinical trials (RCTs). In this paper, we report data from a naturalistic study of successful treatments in clinical practice that bear on the external validity of ESTs for three disorders. Participants were 242 experienced doctoral-level clinicians who reported on their last successfully treated patient seeking treatment for clinically significant depression, panic, or anxiety without panic. Successful treatments typically take substantially longer than the 8-16 sessions characteristic of efficacy trials for these disorders, even for the briefest treatments (cognitive-behavioural). Of particular relevance for generalizability from RCTs, most patients in clinical practice present with multiple problems other than a single Axis I disorder, which clinicians of all theoretical orientations recognize and treat, and these co-occurring conditions have a substantial impact on treatment length in everyday practice. The data suggest the importance of effectiveness research in bridging research and practice. They also point to the utility of distinguishing two complementary ways in which effectiveness research can be understood and implemented: by starting with efficacy trials and then testing treatments with promising results in the laboratory using broader community samples; or by starting with everyday clinical practice, examining patterns of covariation between specific interventions and outcomes at clinically meaningful follow-up intervals with diverse and ecologically valid samples and using these data to generate prototypes of treatments that can be used to guide the next generation of experimental studies.